Paratyphoid Vaccine Development: Benefits Evaluation

Abstract

Introduction: Paratyphoid is suspected to cause millions of deaths annually, and no paratyphoid vaccine exists. Our objective is to assess the epidemiologic data and the relevant pros and cons surrounding the benefits of developing a paratyphoid vaccine.

Material and methods: We performed a critical appraisal of the relevant existing data on paratyphoid (by querying Pubmed®, Embase®, national and supranational databases), and a static economic evaluation comparing strategies with or without implementation of a potential paratyphoid vaccine in India.

Results: The most significant share of paratyphoid burden is borne by the Indian subcontinent (82% of incidence, deaths and DALYs). Children <15 account for ~57.4% and 65.2% of deaths and DALYs. Eighty to 85% of paratyphoid strains reported to CDC and ECDC were probably acquired abroad, mainly by VFRs who had traveled to the Indian subcontinent. In some Indian and Chinese areas, the incidence of paratyphoid has exceeded that of typhoid. Fluoroquinolones resistance is widely spread. Multi-drug resistance remains low. The implementation of a paratyphoid vaccine in India would probably avert tens of thousands deaths and millions of DALYs.

Discussion and recommendations: Paratyphoid epidemiological data are fuzzy. Estimates have wide uncertainty intervals. No trend can be confirmed. The distinction between Salmonella Paratyphi B pathotypes is rarely done. An emergence of MDR or XDR Salmonella Paratyphi could occur, by transfer of these resistances from Salmonella Typhi to Salmonella Paratyphi. Available data do not advocate for a paratyphoid vaccine targeting travelers only. The implementation of a paratyphoid vaccine in India would be probably cost-effective, even cost-saving if targeting children < 15.

Conclusion: The wide uncertainty around paratyphoid epidemiology reflects the paucity of available data and highlights the need for more widespread active surveillance of enteric fevers, with distinction between Salmonella serotypes and pathotypes. A paratyphoid vaccine would probably reduce the burden of enteric fevers in endemic countries. As countries move forward with vaccine roll-out, better surveillance should be a priority.

Keywords: Salmonella, Paratyphi, Paratyphoid, Enteric fever, Vaccine

Introduction

Background

Typhoid and paratyphoid fevers, collectively referred to as enteric fever, are caused by systemic infection with Salmonella enterica serovars Typhi (S. Typhi) and Paratyphi (S. Paratyphi) A, B, and C  (1). These bacteria are collectively referred to as “typhoidal Salmonellae”. S. Paratyphi A and S. Paratyphi B (and uncommonly S. Paratyphi C) cause a clinically indistinguishable disease from an infection by S. Typhi (2,3). Nevertheless, some publications suggest that the prognosis of paratyphoid is milder and symptoms are less severe (4). Humans can

be short- or long-term carriers of S. Typhi and S. Paratyphi.

Transmission of the infection is by the fecal-oral route, through contaminated water or food or, rarely, through person-to-person contact (1). Typhoid and paratyphoid infections are relatively common in countries with poor water supply and sanitation, especially south Asia, some countries in south-east Asia, and sub-Saharan Africa, where they are a major cause of death and disability, especially among children (5). Whereas most non-typhoidal Salmonella spp infections typically induce diarrhoeal illness and less commonly cause bloodstream infection, typhoid and paratyphoid infections produce primarily bacteraemic febrile illnesses, with prolonged high fever, headache, and malaise being characteristic symptoms.

Without effective treatment, typhoid and paratyphoid fevers can lead to altered mental states (termed “the typhoid state” in English and “tuphos” in French), ileus, gastrointestinal bleeding, intestinal perforation, septic shock, and death (1). The optimal antibiotic choice for the treatment of enteric fevers is uncertain. Monotherapy is commonly used, but combination therapies (e.g. ceftriaxone/ciprofloxacin) have also been commonly used (6). In the context of multiple‐drug resistance (MDR) to first line antibiotics, fluoroquinolones were considered the drugs of choice.

However, due to the spread of fluoroquinolone resistance in endemic areas, parenteral therapy with a third generation cephalosporin (C3G), or treatment with azithromycin have become more common (7). Furthermore, between 1% and 5% of patients with acute typhoid infection have been reported to become chronic carriers of S. Typhi in the gallbladder (8), and epidemiologic studies have shown that long-term S. Typhi persistence within the gallbladder results in a dramatically increased likelihood of gallbladder carcinoma (9,10).

In addition, a recent study suggested that other Salmonella serotypes (among which S. Paratyphi) may also play a role in gallbladder carcinogenesis (11). The United Nations (UN) Sustainable Development Goals (SDGs) provide a relevant framework within which control of typhoid and paratyphoid fever, through integration of vaccination with other intervention, should be leveraged and implemented by policy makers (most notably SDG #6: “Ensure availability and sustainable management of water and sanitation for all”). (12)

Recently, the Global Burden of Diseases (GBD) 2017 Typhoid and paratyphoid collaborators published a systematic analysis using models to estimate the incidence of typhoid and paratyphoid (3). For the year 2017, the authors estimated that there have been 10.9 million [9.3–12.6] cases of typhoid fever and 3.4 million [2.7–4.2] cases of paratyphoid fever globally, responsible for estimated 116.8 thousands [65.4 – 187.7] deaths from typhoid and 19.1 thousands [8.7–37.3] from paratyphoid fever. In the past century, a heat inactivated whole-cell parenteral vaccine containing S. Typhi, S. Paratyphi A and S. Paratyphi B (TAB vaccine) has existed, and has been used for several decades.

Although this vaccine is no longer administered due to the severity of its side effects (up to 30% of vaccinees with fever), it attests to the possibility of vaccine-induced protection against paratyphoid (13). Since the appearance of polysaccharide typhoid vaccines in 1994, the TAB vaccine was abandoned and despite a substantial burden from paratyphoid fever, no more vaccine targeting S. Paratyphi exists. Besides, cross-protection provided by existing typhoid vaccines vis-à-vis S. Paratyphi is limited (14).

Objectives

Following Word Health Organization’s (WHO) recommendations (2), Sanofi Pasteur is considering to launch a typhoid conjugate vaccine (TCV) on the market. The primary target market would be travelers and military from Europe and North America. Endemic countries would be a secondary target. As this new typhoid conjugate vaccine could be developed, the question is to know whether it is worthwhile to seize this opportunity to develop a paratyphoid vaccine at the same time (standalone or combined with TCV), rather than a standalone TCV.

Indeed, the WHO Working Group on quality, safety and efficacy of TCV stated in 2013 that TCV should only be a step towards a bivalent vaccine that would be able to prevent both typhoid and paratyphoid disease (15). For this purpose, there is a need to assess epidemiologic burden to confirm the interest to develop a paratyphoid vaccine for traveler market (travelers & military from high income countries [HIC], and emerging countries).

There is also a need to conduct an economic evaluation to assess the cost-effectiveness of the development and implementation of such a vaccine. As a consequence, the main objective of this study is to assess if there is an epidemiological and/or economic interest for the development of a paratyphoid vaccine, primarily for travelers and military from HIC, and secondarily for people living in endemic countries. To achieve this objective, this evaluation involves epidemiological (geographical distribution, incidence, burden of disease), microbiological (antibiotic resistance data), economic (target market size, economic evaluation) assessments.

Material and Methods

This is a critical appraisal of the relevant data in order to define best guess, range of uncertainty and recommendations. This is not an exhaustive review of the existing literature, neither a (rapid) systematic review, nor a focused literature review. This approach is mainly driven by operational constraints (thus, time constraints), as the project involves several departments of Sanofi Pasteur, and the deliverables are due in a defined (limited) time schedule.

For this work, we mostly used epidemiological data from the GBD 2017 Typhoid and Paratyphoid Collaborators publication (Lancet Infect Dis, 2019) (5) and from the Global Health Data Exchange (GHDx) (Institute for Health Metrics and Evaluation [IHME], Seattle / University of Washington, available at http://ghdx.healthdata.org/), which is presented to be “the world’s most comprehensive catalog of surveys, censuses, vital statistics, and other health-related data”.

Additional up-to-date (or most recent) epidemiological data, economic evidence data, and relevant publications concerning antibiotic susceptibility have been retrieved from articles from peer reviewed journals (found by querying (due to time constraint) in 2 of the major existing databases: Pubmed ® and Embase ®), but also from national and supranational public databases or publications (European Centre for Disease Control and prevention [ECDC], United States Centers for Disease Control and prevention [CDC], WHO).

The following search terms were used: ‘typhoid*’, ‘paratyphoid*’, ‘Salmonella Paratyphi’, ‘Salmonella enterica’, ‘enteric fever’, ‘antibiotic resistance’, ‘antibiotic susceptibility’, secondarily with a combination of keywords and terms associated to travel including ‘travel*’, ‘military’, ‘return’, ‘traveller’; ‘traveler’. Publications were not limited in terms of language. For the publications that were not in English, nor French language, a translation was performed using the GoogleTM Translate extension in GoogleTM Chrome.

Due to limited data on paratyphoid epidemiology in travelers, we performed a static economic evaluation based on decision tree comparing a strategy with or without implementation of a potential paratyphoid vaccine in the main endemic country for paratyphoid: India. Country and demographic characteristics were retrieved from the World Bank database (https://data.worldbank.org/country/india), and the epidemiologic parameters from the supplement to the GBD 2017 Typhoid and Paratyphoid Collaborators publication (5) and from the GHDx (IHME, http://ghdx.healthdata.org/).

The different probabilities used in the decision trees, and the costs of treatment and of intervention (vaccine implementation strategy) were from Bilcke et al. (Lancet Infect Dis, 2019) (16), assuming taking typhoid treatment and prevention costs as proxies of paratyphoid ones. We also assumed TCV vaccine efficacy at 46 months (0.85 [0.80 – 0.88]) (17) to be similar with the efficacy of a potential paratyphoid vaccine. For the vaccination strategy, a conservative vaccination coverage rate was arbitrary set at 50%, and an optimistic rate at 84% for the first injection of the vaccine (this rate being the average coverage rate for the vaccines implemented in India for more than 10 years (18)).

Finally, we set the case-fatality rate (CFR) at 0 for outpatients, considering that these were non-severe cases, and that in case of aggravation they were recaptured in hospitalized patients. The inputs used for the economic evaluation were retrieved in the literature and national and supra-national databases, using the same process as described above, and the decision tree model is depicted in Figure 1.

The calculations were done in MicrosoftTM Excel 2013. Figure 1 – Decision tree model with fixed probabilities (in red) used for economic evaluation concerning paratyphoid fever and paratyphoid vaccine strategy in India. Not mentioned probabilities are those which can vary according to the tested strategy A SWOT (Strengths, Weaknesses, Opportunities, and Threats) analysis has been performed as part of the final discussion.

Analysis

Epidemiologic aspects

Paratyphoid is still a major public health issue in LMIC, and especially in the Indian subcontinent, which concentrates at the same time, far ahead of other areas, the hugest burden of the disease in terms of deaths, DALYs and economic costs (Table 3) (24). Furthermore, the part of paratyphoid among enteric fevers is rising in some parts of this area, sometimes overpassing that of typhoid (22). Some hypothesize that the rise of S. Paratyphi A may be due to higher rates of vaccination which protects patients against S. Typhi but not S. Paratyphi, allowing the latter to flourish (34).

However, as discussed before in the limitations chapter, global current and available data on paratyphoid should be considered with caution. Indeed, apart from the uncertainty about the incidence of the disease (due to poor reporting systems), paratyphoid transmission models support a scenario in which the contribution of asymptomatic carriers to overall disease burden is difficult to discern in the context of the high overall number of transmission events occurring in epidemics and highly endemic regions (35).

This put together lead to vast uncertainty intervals in the estimation of the incidence of the disease. Thus, even if there seems to be a trend to decline in the global incidence of paratyphoid through the years, this trend is difficult to confirm as the confidence intervals are overlapping through the years (Figures 2 & 3).

Antimicrobial resistance

The epidemiologic situation is challenged by a rising antibiotic resistance in S. Paratyphi that has already reached 100% of tested strains against fluoroquinolones in certain districts of India (38). Also, an outbreak of extensively drug-resistant (XDR) typhoid (which is resistant to all the five classes of antibiotics usually recommended for the treatment of typhoid fever) is  still going on in Pakistan since 2016 and has now affected more than 5,200 people (44), raising  fears  of  antibiotic  failure  at  global  level (45).

Besides this, it has been shown that plasmids with identical antimicrobial resistance insertions can be transferred from S. Typhi to S. Paratyphi A, including DNA sequences encoding multidrug resistance (46). Thus, the emergence of MDR S. Paratyphi at highest levels and, worse, of XDR S. Paratyphi in the future cannot be dismissed.

Case of travelers

With most of the paratyphoid cases observed in the USA, the EU/EEA, Israel and Australia  occurring in VFRs returning from the Indian subcontinent (India, Pakistan, Bangladesh), among which a majority of S. Paratyphi A strains with fluoroquinolone resistance, there is a perfect alignment of the epidemiology of paratyphoid fever among travelers, with global epidemiology.

The seasonal pattern observed in the EU/EEA (Figure 6), with peaks of paratyphoid cases in September and late spring, most probably reflects travel during holiday periods, with disease onset after returning home. It is also interesting to notice that paratyphoid fever cases also follow a seasonal pattern in Asia, with a peak season from May–October (34). However, the incidence of the disease in HIC travelers remains extremely low (2,579 paratyphoid strains reported to ECDC in 6 years, 1,756 to CDC in 7 years) compared to the estimated 3.4 million worldwide annual paratyphoid cases (5).

There is probably a lack of exhaustivity in the reporting systems (even though paratyphoid reporting is mandatory both in the USA and the EU/EAA), as confirmed in reports (28–30), leading to underestimation of the incidence in the traveler population. Furthermore, there is a dramatic discrepancy between the hospitalization rates of paratyphoid cases in India (6% [2-15]) and USA (67.8%), probably explained by socio-economic parameters.

Economic evaluation

In 2003, Hutubessy et al., writing on behalf of WHO’s Choosing Interventions that are Cost–Effective project (WHO-CHOICE) suggested that “interventions that avert one DALY for less than average per capita income for a given country (…) are considered very cost–effective; interventions that cost less than three times average per capita income per DALY averted are still considered cost–effective; and those that exceed this level are considered not cost–effective” (47).

Based on these thresholds, the results of our economic evaluation suggest that, in addition to averting tens of thousands of deaths and several million DALYs per year, the implementation of a paratyphoid conjugate vaccine could be very cost-effective, no matter which of the four scenarios studied is selected (Table 6), but a scenario in which the target would be children under 15 years old (because of poor data on < 2 years old children demography in India) would be both very cost-effective and cost saving, even in the hypothesis of a vaccine coverage rate at only 50% (Table 6).

This is aligned with the now admitted fact that immunization is one of the most cost-effective ways to save lives and improve health, and that “immunizing children is one of public health’s ‘best buys’” as stated by the Global Alliance for Vaccines and Immunization (GAVI) (48,49) Nevertheless, the results of our economic evaluation should be taken with caution because of some sensitivity limitations due to uncertainties in several input parameters. The first one is the use of input proxies related to typhoid disease and TCV for the disease management probabilities and the direct and intervention costs, because of the lack of paratyphoid specific data.

Secondly, since a paratyphoid vaccine does not exist yet, we assumed the vaccine efficacy of a potential paratyphoid vaccine to be similar to that of TCV. Thirdly, our model predicted, at basal level (without vaccine implementation), 107,962 annual deaths due to paratyphoid in India, when the most recent publications estimate 13,849 [6,467 – 27,038] annual deaths in that country (5,24).

This may due to the conjunction of the paucity of paratyphoid mortality data making the use of models necessary to estimate the mortality, and the fact that these publications do not clearly take into account the mortality in people who do not seek health care and have probably a higher probability of death in the absence of treatment (41). That said, we tested the robustness of our model by implementing a critical scenario with a vaccine efficacy at 50%, a vaccine coverage rate at 50% and we doubled the vaccine procurement cost (3.00 USD instead of 1.50 USD). We found out that, even with this pessimistic scenario, the implementation would be still very cost-effective regarding the threshold cited above.

SWOT analysis

Strengths – Sanofi Pasteur expertise (conjugation, bivalent vaccines) – Development & production sites in India (Shantha Biotech, Indian vaccine manufacturer, 100% owned by Sanofi); Weaknesses – Development costs – Foreseeable difficulties in research and development?

External (to Sanofi Pasteur) Opportunities – Indirectly at WHO agenda (12) – Burden of paratyphoid more and more assessed in endemic countries – Paratyphoid incidence is progressively exceeding that of typhoid in endemic countries (22,23) – Endemic disease in regions visited by HIC emerging countries travelers and military – Prevention of the rise of antibiotic resistance (existing fluoroquinolone resistance, and foreseeable MDR and XDR) – Participation to the prevention of gallbladder cancer (to be assessed) Threats – Existing competition, particularly the oral live-attenuated vaccine Ty21a (Vivotif® by PaxVax, Inc), with confirmed efficacy against S. Paratyphi B (49%), and alleged protection against S. Paratyphi A (to be proven) (13) – Limited incidence and burden on HIC travelers (size of the market).

Recommendations

This study mainly highlighted the lack of accurate available data on paratyphoid, as only a few, small, poorly designed studies have addressed the disease epidemiology and economic features concerning paratyphoid. However, using the available data (essentially model-based, and with their intrinsic uncertainty) and proxies based on the accumulated scientific and economic knowledge on typhoid and its prevention, our model found out that the implementation of a paratyphoid conjugate vaccine in the most impacted country (India) would probably be very cost-effective (particularly if targeting children under five), and avert tens of thousands of deaths and millions of DALYs.

The opportunity window created by the recent inversion of the S.Typhi/S. Paratyphi incidence ratio in endemic countries and the possible emergence of MDR and XDR strains in these regions as a sword of Damocles on these latter, should be an incentive for vaccine manufacturers (among which Sanofi Pasteur) to develop a paratyphoid vaccine (Table 7). This opportunity has already been seized by manufacturers (e.g. Bharat Biotech, Biological E), as well as public funded institutes (International vaccine Institute, Indian National Institute of Cholera and Enteric Diseases..), which are trying to develop a S. Paratyphi A vaccine, with the intention to target endemic countries (50–52).

From Martin LB et al. (Vaccine, 2016) (50) Yet, existing epidemiologic data do not support a scenario in which the only business target would be travelers and military, and further accurate data on travelers are needed to clearly identify the risk and the market. Still, the eventuality to target only HIC travelers and military would emphasize ethical concerns that already exist for typhoid vaccines: this is the typhoid vaccines paradox: the vaccines have been evaluated among populations in endemic (LMIC) countries, and have been predominantly used among travelers from HIC and military.

There is no guarantee that what has been accepted in the 1990s when the polysaccharid typhoid vaccines were launched will be accepted in the 2020’s. In that hypothesis, Sanofi Pasteur has a head start with Shantha Biotech (an Indian vaccine manufacturer, based in India, and 100% owned by Sanofi), which may allow the company to be present both in the endemic countries market (by limiting the production and distribution costs and participating to GAVI and WHO prequalification calls) and the travelers market.

Conclusion

To date, it remains difficult to make evidence-based decisions on paratyphoid fever prevention, as only a few, small, poorly designed studies have addressed disease epidemiology and economic features. Nevertheless, it is fairly well established that the vast majority of paratyphoid cases occur in South Asia (mostly in three countries of the Indian subcontinent: India, Pakistan and Bangladesh), Southeast Asia, and sub-Saharan Africa. These regions, known to be popular destinations for travelers from high-income countries and for military of the Western armies, also concentrate the vast majority of typhoid cases and also the heaviest global burden of both diseases.

However, no trend in the incidence of paratyphoid worldwide can be firmly confirmed. Typhoid is still responsible for most of the enteric fever burden in countries in the above-mentioned areas, in terms of deaths as well as in term of disabilities, but in parallel the paratyphoid/typhoid ratio is reversing in endemic countries, at such a point that in some Indian and Chinese areas, the incidence of paratyphoid has exceeded that of typhoid, with a tendency to develop antimicrobial resistance (mainly fluoroquinolone resistance, and MDR at lower levels).

But due to genotypic similarities between S. Typhi and S. Paratyphi, the emergence of MDR or XDR S. Paratyphi in the future cannot be excluded. With a probable high cost-effectiveness of the implementation of a future paratyphoid vaccine in India, along with other pros like the possibility to avert tens of thousands of deaths, millions of DALYs, and also possibly numerous gallbladder cancers, the results of our study are in favor of the development of a paratyphoid vaccine, and its implementation in endemic countries.

Nevertheless, more accurate and comprehensive data are needed to improve the reliability of our study and model in order to answer the question of cost-effectiveness of a paratyphoid vaccine targeting only travelers from high income and emerging countries and military. In order to fill the data gaps, future surveys and publications about enteric fevers, as well as national and supranational databases on the subject and microbiological laboratories, should clearly consider paratyphoid and typhoid separately and, among S. Paratyphi strains, clearly consider S. Paratyphi B ‘d-tartrate negative’ and S. Paratyphi B ‘d-tartrate positive’. These simple tips would highly help to improve the appraisal of the real incidence and burden of paratyphoid.